无需标签信息，单轮平均梯度中还原图像细节
MAGIA: Sensing Per-Image Signals from Single-Round Averaged Gradients for Label-Inference-Free Gradient Inversion
- 通过随机子集探测感知每张图的隐含信号
- 大批次下实现高保真多图重建，超越现有方法
- 无需额外数据，计算开销与常规求解器相当
我们研究在单轮平均梯度（SAG）场景下的梯度反演问题，该场景中单个批量均值梯度内嵌了每个样本的线索。本文提出MAGIA，一种基于动量的自适应修正框架，无需标签推断即可从梯度中感知每张图像的潜在信号。MAGIA的核心创新包括：1）闭式组合重缩放，提供可证明更紧的优化边界；2）全批量与子集损失的动量混合机制，保障重建鲁棒性。大量实验表明，MAGIA在大批量场景下显著优于先进方法，实现高保真多图重建，且计算开销与标准求解器相当，无需任何辅助信息。
原文摘要 · Abstract (English)
We study gradient inversion in the challenging single round averaged gradient SAG regime where per sample cues are entangled within a single batch mean gradient. We introduce MAGIA a momentum based adaptive correction on gradient inversion attack a novel label inference free framework that senses latent per image signals by probing random data subsets. MAGIA objective integrates two core innovations 1 a closed form combinatorial rescaling that creates a provably tighter optimization bound and 2 a momentum based mixing of whole batch and subset losses to ensure reconstruction robustness. Extensive experiments demonstrate that MAGIA significantly outperforms advanced methods achieving high fidelity multi image reconstruction in large batch scenarios where prior works fail. This is all accomplished with a computational footprint comparable to standard solvers and without requiring any auxiliary information.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。